Diazepam autoinjector intramuscular delivery system versus diazepam rectal gel: A pharmacokinetic comparison

Epilepsy Research (2011) 93, 11—16

journal homepage: www.elsevier.com/locate/epilepsyres

Diazepam autoinjector intramuscular delivery system versus diazepam rectal gel: A pharmacokinetic comparison William R. Garnett a,∗, William H. Barr b,1, Leslie E. Edinboro c,2, H. Thomas Karnes d,3, Mike Mesa e,4, Gerald L. Wannarka f,5 a

Center for Drug Studies, Departments of Pharmacy and Neuroloy, Smith Building Room 334, Virginia Commonwealth University, School of Pharmacy, 410 North 12th Street, PO Box 980583, Richmond, VA 23298-0583, United States b Center for Drug Studies, Department of Pharmacy, Smith Building Room 409, Virginia Commonwealth University, School of Pharmacy, 410 North 12th Street, PO Box 980583, Richmond, VA 23298-0583, United States c Quest Diagnostics, Nichols Institute, 14225 Newbrook Drive, Chantilly Virginia 20151, United States d Virginia Commonwealth University, School of Pharmacy, 410 North 12th Street, PO Box 980533, Richmond, VA 23298-0533, United States e Meridian Medical Technologies, 6350 Stevens Forest Road, Suite 301, Columbia, MD 21046, United States f Wannarka Consulting, 3284 Danmark Drive, Glenwood, MD 21738, United States Received 30 January 2009; received in revised form 29 September 2010; accepted 3 October 2010 Available online 9 December 2010

KEYWORDS Diazepam; Epilepsy; Intramuscular; Pharmacokinetics; Rectal; Seizure

Summary Rectal administration of diazepam (DZ) has been used effectively in patients with epilepsy who experience acute repetitive seizures, but rectal gel may be difficult to administer during a seizure and is subject to variable drug absorption. An intramuscular (IM) autoinjector DZ formulation may offer a practical alternative to rectal DZ. This single-center, open-label, 3treatment, 3-period crossover study compared the pharmacokinetic and safety profiles of 10 mg DZ administered rectally in 24 healthy, fasted and fed subjects versus IM autoinjector delivery in fasted subjects. Blood samples were collected at baseline and for up to 24 h postdose and plasma DZ concentrations were determined by liquid chromatography and tandem mass spectrometry.

∗ Corresponding author at: Center for Drug Studies, Departments of Pharmacy and Neurology, Smith Building Room 334, Virginia Commonwealth University, School of Pharmacy, 410 North 12th Street, PO Box 980583, Richmond, VA 23298-0583, United States. Tel.: +1 804 828 8328; fax: +1 804 828 8359. E-mail addresses: [email protected] (W.R. Garnett), [email protected] (W.H. Barr), [email protected] (L.E. Edinboro), [email protected] (H.T. Karnes), [email protected] (M. Mesa), [email protected] (G.L. Wannarka). 1 Tel.: +1 804 828 8334; fax: +1 804 828 6902. 2 Tel.: +1 703 802 7217; fax: +1 703 802 7041. 3 Tel.: +1 804 828 3819; fax: +1 804 828 8359. 4 Tel.: +1 443 259 7800; fax: +1 443 256 7801. 5 Tel.: +1 443 266 7170; fax: +1 443 266 7170.

0920-1211/$ — see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2010.10.001

12

W.R. Garnett et al. There were no significant differences between plasma concentrations for rectal administration of DZ in fasted and fed subjects at any time point. Intramuscular DZ administration resulted in more rapid and less variable drug absorption than rectal delivery. At 30 min postdose and at all subsequent evaluations, IM administration resulted in significantly higher areas under the curve versus rectal administration in fasted subjects (p < 0.05). This significant difference was sustained throughout the remainder of the 24-h study period (p < 0.05). All reported adverse events were considered mild, and none required treatment. © 2010 Elsevier B.V. All rights reserved.

Introduction Epilepsy is a chronic neurologic disorder that has been estimated to affect approximately 2.5 million Americans (Centers for Disease Control and Prevention, 2010). Some people with epilepsy have, in addition to their usual seizure pattern, repetitive exacerbations, which may last minutes or hours and have a pattern that is different from the patient’s usual seizure pattern. These exacerbations are commonly referred to as acute repetitive seizures (ARS) (Mitchell, 1996). Benzodiazepines (diazepam [DZ] and lorazepam) are effective for many seizure types, have a rapid onset of action with a good safety profile, and are widely used for the management of patients with ARS (Bone, 1993; Mitchell, 1996; Kapur, 2002). Although a number of benzodiazepine formulations — including oral and sublingual tablets and rectal suppositories and gels — have been used to treat ARS (Garofalo et al., 1989; Greenblatt et al., 1982; Milligan et al., 1983; Yager and Seshia, 1988; Lombroso, 1989; Vanquix, 2005; Schroeder et al., 1996; Dreifuss et al., 1998), they have various limitations (Graves et al., 1987). For example, oral and sublingual formulations can be difficult or hazardous to administer, and high interpatient variability in drug absorption after rectal administration has been observed (Dhillon et al., 1982a,b; Graves et al., 1987; Lombroso, 1989; Dreifuss et al., 1998). Results from one study of DZ rectal suppositories in pediatric patients between the ages of 1 and 15 years indicated plasma concentrations 10 min after dosing that ranged from 26 to 121 ng/ml (Dhillon et al., 1982b). Similarly, a study of rectal DZ solution in adults with epilepsy indicated maximum plasma concentrations (Cmax ) ranging from 215 to 415 ng/ml, and time to Cmax (Tmax ) ranging from 13 to 60 min (Dhillon et al., 1982a). Rectal administration of DZ gel (Diastat® , Diastat® AcuDialTM ) also has a long Tmax (1.5 h) and variable clearance with a coefficient of variation (CV) for elimination half-life (t1/2 ) of 43% (Diastat, 2006). Intramuscular (IM) DZ administration with a needle and syringe has also been used to treat ARS, but resulted in significantly slower absorption than rectal administration (Kanto, 1975; Magnussen et al., 1979; Rey et al., 1999). An IM formulation of DZ administered by autoinjector may offer a practical alternative that addresses the limitations of oral, sublingual, rectal, and previous IM preparations. This study compared the pharmacokinetics, adverse events (AEs), and administration-route preference for DZ rectal gel among fed and fasted subjects with the DZ IM autoinjector. The comparison between IM and rectal formulations was carried out under ‘ideal’ fasting conditions, in which an empty rectal vault might be expected to increase DZ absorption, and fed

conditions that might more accurately reflect ‘real’ clinical use.

Methods Study design This was a single-center, open-label, 3-treatment, 3-period crossover study, with a 7-day washout period after each treatment. A total of 24 healthy, normal subjects (22 men and 2 women) were randomized to receive 10 mg of DZ under 3 conditions: ‘ideal’ rectal (IR), ‘real’ rectal (RR), and IM autoinjector (Meridian Medical Technologies, Inc.). ‘Ideal’ rectal administration involved a single administration of DZ after a cleansing enema the evening prior to dosing, an overnight fast, and another cleansing enema in the morning prior to drug administration. ‘Real’ rectal administration involved a single administration of DZ after a standard evening meal and bedtime snack on the day before dosing, and at 2 h after a standard breakfast on the day of dosing. Intramuscular autoinjector administration involved a single injection of DZ into the midanterolateral thigh using the autoinjector device. Subjects receiving the IM autoinjector followed the fasting requirements for IR administration, but did not receive the cleansing enemas. Before the trial was initiated, the protocol was reviewed and approved by an institutional review board (IRB), and an IRB-approved, written informed consent was obtained from all subjects.

Subjects Men and women 18—55 years of age, inclusive, who were in general good health based on medical history and had clinically acceptable results on all screening assessments were eligible to participate. Women could not be pregnant or lactating, and they had to be surgically sterile, postmenopausal, or practicing an acceptable method of birth control for at least 3 months prior to the start of the study. All subjects were required to be within 25% of normal body weight, as determined by the 1996 Metropolitan Life Tables. Subjects were excluded if they had a current acute or chronic disease or a history of a clinically important medical disorder; a history of anaphylaxis, a documented hypersensitivity reaction, or a clinically important idiosyncratic reaction to DZ or other benzodiazepines; any condition that could interfere with the absorption, distribution, metabolism, or excretion of drugs, or any condition that might confound the analyses; or a history of human immunodeficiency virus infection/acquired immune deficiency syndrome, viral hepatitis (other than hepatitis A), or laboratory-confirmed obstructive sleep apnea. Subjects were also ineligible if they had donated blood or plasma or received another investigational drug within 30 days of the study, or if they had a history of alcohol abuse, a history or current evidence of abuse of licit or illicit drug substances, or a positive urine drug screen for drugs of abuse. Any subject who used an over-the-counter (OTC) medication or was expected to use any prescription or OTC medication within 72 h of dosing that might interfere with the evaluation of study medication was excluded.

Diazepam autoinjector intramuscular delivery system versus diazepam rectal gel Drug administration All subjects were admitted to the Center for Drug Studies (CDS) on 3 separate occasions. On each occasion, they entered the CDS on the day before receiving the dose of DZ and stayed until after the 24-h blood sample had been collected and they were deemed physically able to go home. All subjects refrained from strenuous activity while in the CDS. Due to the small sample size, only 3 sequences with 8 subjects randomized to each sequence were used. The IR, RR, and IM treatments were randomly assigned in 1 of the 3 sequences: (i) IR, RR, IM; (ii) RR, IM, IR; or (iii) IM, IR, RR. Since IR, RR, or IM administration was equally likely to be first, second, or third in these 3 sequences, they were balanced but did not represent all possible sequences. The study was conducted in 2 groups of 12 subjects to accommodate the complexity of the study tasks. Thus, there were 6 periods, each with 4 subjects. The 2 groups were not intermingled. An analysis of variance (ANOVA) indicated no statistically significant difference between the 2 groups. A nurse administered the rectal gel (Diastat® ) according to instructions for caregivers included in the product packaging. If the subject expelled the rectal DZ within the first 30 min due to defecation or oozing, the amount expelled was estimated but the dose was not repeated. The IM injection was administered to a cleansed area on the midanterolateral thigh. The DZ autoinjector is a prefilled, nose pressure—activated system that is similar in design to the approved military 10-mg diazepam Convulsant Antidote Nerve Agent Autoinjector (CANA). The DZ autoinjector utilizes the same drug formula, depth of injection, speed of injection, and needle gauge as the CANA. With the DZ autoinjector, an internal spring delivers a sterile 2-ml dose of 5 mg/ml DZ for IM injection. Removing the safety release and pressing the nose of the injector against the injection site with approximately 4 pounds of pressure triggers ejection of an 18- to 23-mm 20-gauge needle and provides adequately deep IM delivery of DZ in a few seconds. To ensure appropriate DZ dosing in study subjects, the IM autoinjector was held in place at depth while the nurse counted to 10. The same nurse, who was trained in drug administration, gave all study drug doses to all subjects. Following administration, all subjects were required to lie on their side for 30 min and abstain from eating for 2 h.

Data collection Plasma samples (10 ml) were collected into heparinized tubes at baseline; at 5, 10, 15, 20, 25, 30, 45, and 60 min; and at 1.5, 2, 4, 6, 8, 12, and 24 h postdose. During the first 30 min after rectal drug administration, subjects were also questioned and assessed to determine if they defecated. After the last blood sample in the third treatment period, subjects were asked which route of DZ administration they preferred. Subjects scored discomfort and pain on a scale of 0—5 (lowest to highest) at 30 and 60 min postdose. The nurse who had administered all study drug doses rated the difficulty of administration on a scale of 0—5 (lowest to highest). Safety was evaluated by AE assessments (at each sample collection time), clinical laboratory tests, vital sign measurements, electrocardiogram, oxygen saturation by pulse oximetry, physical examinations, discomfort and pain ratings, medical history, and concomitant medications.

Analysis of diazepam and desmethyldiazepam in human plasma The analysis of DZ and desmethyldiazepam (DDZ) in human plasma was performed using a validated 1-step, liquid—liquid extraction and liquid chromatography using a tandem mass spectrometry detector (LC/MS/MS). The method was linear from 2.0 to 500 ng/ml for both DZ and DDZ; 2.0 ng/ml was the limit of quantitation (LOQ)

13

for DZ and DDZ. Precision (% relative standard deviation) and accuracy (% difference from nominal) at the LOQ for DZ were 16.1% and 0.0%, respectively. DDZ precision was 5.4% with a corresponding accuracy of −1.9%.

Pharmacokinetic analysis Diazepam has a long t1/2 (20—100 h for DZ and 36—200 h for DDZ) (Diastat, 2006), and in this study there appeared to have been biliary recycling (as has been reported previously) (Vajda et al., 1986). The long t1/2 caused a measurable residual plasma concentration in some subjects at Time 0 on Day 1 of Treatment Periods 2 and 3. A much longer washout period would have been needed to eliminate the residual drug at baseline in the subsequent periods. As a result, the mean area under the curve (AUC) values were slightly higher in Periods 2 and 3 than in Period 1 for all treatment conditions. This produced a statistically significant effect in the sequence term in the ANOVA. Because a Latin square-type randomization scheme was used, each treatment was equally represented in all 3 periods and sequences, and there was no statistically significant effect on the sequence x treatment term in the analysis. Thus, the carryover effect did not significantly impact evaluation of differences between treatments. An effort was also made to pharmacokinetically correct for the residual concentration (Co), based on the assumption that the Co present at the beginning of Periods 2 and 3 would decline in the same manner as the drug that was administered during that period. Therefore, a correction for the Co at each successive time point, or Cr(T), was calculated as Cr(T) = Coe−ˇt , where ˇ was the terminal elimination rate constant corresponding to the long t1/2 that was obtained from the slope of the terminal log-linear concentration—time curve. These values were calculated for each treatment that had a positive Co value (residual at zero time). The Cr(T) value of each time point was used to calculate the areas under the residual concentration curve at successive times for each treatment. For the contribution to the AUC at any given time interval, T (e.g., between 10 min [T1] and 20 min [T2]), the residual contribution AUCr (T1 − T2) was obtained by the trapezoidal rule: AUCr (T1 − T2) = ½ (T2 − T1) (Cr(T1) + Cr(T2)). The residual AUC values were subtracted from the original (uncorrected) values to obtain the corrected AUC values. Examination of the individual log plots revealed two problems. First, it was difficult to estimate the terminal slope because the plasma levels were actually in a cyclic mode rather than a true exponential decline. Second, the terminal slope was not the same as the early declining phase. The plasma concentrations showed a significant distributive ˛ phase in the first few hours. Accordingly, a separate correction was also made at successive time intervals as Co(t) = Coe−˛t , where ˛ is the exponential term representing the more rapid plasma concentration decline in the distributive phase. The ˛ values were obtained by fitting a 2-compartment model (WinNonLin® ) to the plasma concentration—time curves. The AUC corrections were performed as described above for the ˇ correction.

Statistical comparisons Examination of the ANOVA sources of variability identified the subject term as the greatest single contributor to the error term. Accordingly, the data were analyzed by two procedures: in the first procedure (A), subject variability was included in the statistical model, rather than the error term; in the second procedure (B), subject variability was included in the error term, rather than the statistical model. Both statistical models were used to evaluate: (i) AUC values, not corrected for the carryover residual concentration; (ii) residual corrected data using the pharmacokinetic term ˛; and (iii) residual corrected data using the pharmacokinetic term

14 Table 1

W.R. Garnett et al. Subject demographics (N = 24).

Sex (n [%]) Male Female Race/ethnicity (n [%]) Black White Age, years (mean [median]) Height, cm (mean [median]) Weight, kg (mean [median]) Body frame (n [%]) Large Medium Small

22(92) 2(8) 11(46) 13(54) 33.9 (34.0) 176.8 (177.8) 77.4 (79.0) 3 (12.5) 12 (50.0) 9 (37.5)

ˇ. The ˛ and ˇ residual corrections gave the same results as the data without residual corrections. The corrections did not prevent a sequence effect at the earlier time points, but the sequence x treatment term was not significant at the 0.05 level for any of the AUC values at any time point.

Results A total of 24 subjects (22 men and 2 women) ranging in age from 19 to 52 years participated in and completed the study. There were no dropouts or subjects removed for protocol violation. Table 1 summarizes subject characteristics.

Pharmacokinetics The mean plasma concentration—time curves presented in Fig. 1 show that there were no substantial differences between the IR and RR conditions at any time point over the first 4 h after administration. There were no statistical differences among the 3 treatments for mean AUC values at 10 and 20 min after dosing (Table 2). At 30 min postdose, IM administration resulted in a mean AUC value that was significantly higher than that for IR administration (p < 0.05). This trend continued at 60 min and at 24 h postdose. Results summarized in Table 2 also indicate similar AUC values for IR and RR DZ administration over all time periods evaluated.

Figure 1 Mean plasma concentration—time curves for IR, RR, and IM DZ.

Individual pharmacokinetic values, particularly at the earlier time points, showed a high degree of variability with IR and RR DZ administration. At 10 min postdose (AUC10) for example, values for IR administration ranged from 6 to 1960 ng h/ml and those for RR treatment ranged from 8 to 1232 ng h/ml. Coefficients of variation (CV) at this time point were 70.6% and 52.6%, respectively, for IR and RR treatment. The range of AUC10 values for IM autoinjector treatment was smaller (165—977 ng·h/ml), yielding a 44.6% CV. At all time points from 10 min through 24 h postdose, the CVs were higher for IR and RR treatment than for IM drug administration for AUC (Table 2). Highest peak plasma concentrations were achieved with IM drug administration in 19 of 24 subjects, whereas lowest peak concentrations occurred for 10 subjects with IR administration and for 12 subjects with RR treatment. Some subjects experienced drug expulsion or defecation within 60 min postdose during IR or RR treatment. Over the course of the study, 8 subjects experienced a total of 10 episodes of ‘oozing’ (<1.0 ml) after rectal administration of DZ. It was not possible to determine if the material expelled was the surgical lubricant given prior to administration or the DZ gel. These episodes of oozing did not alter the DZ concentration—time curves in a consistent manner. Of the 4 subjects who experienced oozing with RR treatment, 1 showed a lower DZ concentration (0—4 h) than after either IR or IM treatment, 1 was similar to IR but lower than IM and 1 was similar to both IR and IM. Of the 2 subjects who showed only IR oozing, 1 showed a lower DZ concentration than after IR or IM treatments and the other showed a DZ concentration higher than IR but similar to IM. For the 2 subjects who showed oozing following both IR and RR treatments, both had lower DZ concentrations than following IM treatment and both had lower DZ concentrations following RR than IR treatments. One subject had a bowel movement at 51 min after dosing with no gel noted; this did not alter the concentration—time curve.

Adverse events All reported AEs were considered mild, and none required treatment. A total of 12 subjects complained of an ache, pain, burning, or heaviness in the right leg following IM injection, and 4 subjects complained of a burning or tingling sensation in the rectum following rectal administration of DZ. All of these complaints resolved spontaneously without treatment. Most of the other reported AEs, such as drowsiness or sedation, were anticipated effects of DZ and resolved without treatment. There were no serious AEs and no subject was removed from the study because of an AE. There were no changes in clinical laboratory tests, electrocardiograms, respiratory function (as reflected by oxygen saturation), or physical examination results during any treatment period. Other than 1 subject who reported mild hypotension, there were no changes in vital sign measurements. This individual required no medication or further treatment and completed the study as scheduled. There was no significant difference in the rate of nonserious central nervous system (CNS) AEs during IR and RR treatment (66.7% versus 87.5%) or during RR and IM treatment (87.5% versus 95.8%). However, significantly more

Diazepam autoinjector intramuscular delivery system versus diazepam rectal gel Table 2

15

Mean (% CV) pharmacokinetic parameters for IR, RR, and IM DZ.

AUC10 (ng·h/ml) AUC20 (ng·h/ml) AUC30 (ng·h/ml) AUC60 (ng·h/ml) AUC24 (ng·h/ml) Cmax (ng/ml) Tmax (min)

IR

RR

IM

618.3 (70.6) 2103.5 (53.8) 3828.4 (50.6) 9332.5 (46.9) 142078 (36.6) 233.9 (41.6) 84 (168)

608.8 (52.6) 2187.3 (43.7) 3944.3 (43.9) 9173.6 (46.6) 122692 (41.9) 209.1 (44.6) 36 (76)

474.7 (44.6) 2212.6 (34.2) 4695.0* (31.1) 12799.8*,† (25.8) 165085* (22.4) 303.9 (23.6) 52 (62)

AUC10 = area under the plasma concentration—time curve (AUC) from 0 to 10 min; AUC20 = AUC from 0 to 20 min; AUC30 = AUC from 0 to 30 min; AUC60 = AUC from 0 to 60 min; AUC24 = AUC from 0 to 24 h; Cmax = maximum plasma concentration; DZ = diazepam; IM = intramuscular; IR = ‘ideal’ rectal; RR = ‘real’ rectal; Tmax = time to maximum plasma concentration. * p < 0.05 versus IR. † p < 0.05 versus RR.

subjects experienced nonserious CNS AEs during IM treatment (95.8%) versus IR administration (66.7%) (p < 0.01). The most common CNS AEs over all treatment conditions were fatigue and somnolence. Subjects rated pain and discomfort scores on a scale of 0—5 at 30 and 60 min after each dose. Discomfort and pain were reported more often with IM treatments (96% and 92% of subjects at 30 min) versus IR (63% and 38%) or RR (42% and 8%) drug treatments. At 60 min discomfort and pain were experienced by fewer subjects: 79% and 54% of subjects following IM treatments compared with 33% and 17% following IR treatments and 17% and 8% following RR treatments. Both discomfort and pain scores decreased with time (30 and 60 min postdose) in all groups. Discomfort scores decreased from 2.38 to 1.42 (IM group), from 1.13 to 0.53 (IR group) and from 0.79 to 0.25 (RR group). Pain scores decreased from 2.38 to 0.88 in the IM group and from 0.5 to 0.25 in the IR group; the change was negligible (0.08) in the RR group.

Formulation preference A substantial majority of study participants, as well as the nurse administering the study treatments, favored the IM route over rectal administration. A total of 71% of subjects preferred the IM to rectal administration, while 29% of subjects preferred rectal over IM administration. The nurse administering the study medication reported the difficulty of administration as 0.58, 0.50, and 0.04 for the IR, RR, and IM treatments, respectively.

Discussion In this study, the IM autoinjector provided rapid and more reliable absorption of DZ in fasted subjects than in fasted or fed subjects who received rectally administered DZ. These results support the conclusion that the IM autoinjector may overcome limitations of existing approaches to DZ delivery for management of ARS (Tuttle, 1977; Magnussen et al., 1979). The IM autoinjector provided deep, localized delivery of DZ that appeared to differ significantly from the delivery of rectal DZ administration. While the IM autoinjector may provide important advantages for management of many patients with epilepsy and ARS, it has not yet been evaluated in some populations, such as infants and the elderly.

Both methods of rectal administration appeared to have greater variability in absorption than the IM route at all time points. Cleansing the bowel to maximize absorption did not make a significant difference in the total amount of DZ absorbed. It did, however, appear to increase the variability in rectal absorption. The differences in absorption between the IR and RR routes may have been due to differences in the contents of the rectal and sigmoid colon. There are potentially important clinical implications to the delayed or incomplete absorption of rectal DZ. If a caregiver administers rectal DZ and the ARS do not abate, it is likely that a second dose of DZ will be given (Garr et al., 1999). As time passes, more active drug from the initial dose may be absorbed, potentially doubling the amount of active drug administered. There is evidence that DZ accumulates with repeated dosing and that this may result in toxicity (Walker et al., 1998). The greater consistency of absorption seen with the IM autoinjector over IR or RR administration may enhance the predictability of DZ. When queried, nearly three-quarters of subjects preferred DZ via IM autoinjector to rectal DZ, as did the nurse administering the treatments. The preference for the IM autoinjector, despite slightly higher pain and discomfort scores, demonstrates that ease of use and convenience may be more important determinants of patient satisfaction than pain or discomfort, and illustrates the need for an alternative DZ administration that is fast and easy. Because it produced rapid and reliable absorption of DZ, the IM autoinjector appears to be a clinically useful method for patients, medical professionals, emergency medical technicians, and caregivers who treat patients with ARS.

Conflict of interest Dr. William R. Garnett has served as a consultant to Meridian Medical Technologies. Dr. William H. Barr has nothing to disclose. Dr. Leslie E. Edinboro is an employee of Quest Diagnostics Nichols Institute. Dr. H. Thomas Karnes has nothing to disclose. Dr. Mike Mesa is an employee of King Pharmaceuticals, Inc. Dr. Gerald L. Wannaraka was an employee of King Pharmaceuticals, Inc., serves as a part-time consultant to King Pharmaceuticals, Inc., and currently owns stock in King Pharmaceuticals, Inc.

16

Acknowledgements Writing and editorial support for this manuscript was provided by Quintiles Medical Communications, Parsippany, NJ, USA and Flaum Communications, New York, NY, USA. Funding for writing and editorial support was provided by King Pharmaceuticals® , Inc. Funding of this study was provided by Meridian Medical Technologies.

References Bone, R.C., 1993. Treatment of convulsive status epilepticus. JAMA 270, 854—859. Centers for Disease Control and Prevention, 2010. Targeting Epilepsy: Improving the Lives of People with One of the Nation’s Most Common Neurological Conditions. Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Atlanta, GA. Dhillon, S., Ngwane, E., Richens, A., 1982a. Rectal absorption of diazepam in epileptic children. Arch. Dis. Child. 57, 264—267. Dhillon, S., Oxley, J., Richens, A., 1982b. Bioavailability of diazepam after intravenous, oral and rectal administration in adult epileptic patients. Br. J. Clin. Pharmacol. 13, 427—432. November 2006. Diastat [package insert]. Valeant Pharmaceuticals International, Aliso Viejo, CA. Dreifuss, F.E., Rosman, N.P., Cloyd, J., Pellock, J.M., Kuzniecky, R., Lo, W.D., Matsuo, F., Sharp, G.B., Conry, J.A., Bergen, D.C., Bell, W.E., 1998. A comparison of rectal diazepam gel and placebo for acute repetitive seizures. N. Engl. J. Med. 338, 1869—1875. Garofalo, E.A., Hirschorn, K.A., Komarynski, M.A., 1989. Improved control of seizure clusters with rectal diazepam and lorazepam. Cleve. Clin. J. Med. 56 (Suppl. 2), S-277. Garr, R.E., Appleton, R.E., Robson, W.J., Molyneux, E.M., 1999. Children presenting with convulsions (including status epilepticus) to a paediatric accident and emergency department: an audit of a treatment protocol. Dev. Med. Child Neurol. 41, 44—47. Graves, N.M., Kriel, R.L., Jones-Saete, C., 1987. Bioavailability of rectally administered lorazepam. Clin. Neuropharmacol. 10, 555—559.

W.R. Garnett et al. Greenblatt, D.J., Divoll, M., Harmatz, J.S., Shader, R.I., 1982. Pharmacokinetic comparison of sublingual lorazepam with intravenous, intramuscular, and oral lorazepam. J. Pharm. Sci. 71, 248—252. Kanto, J., 1975. Plasma concentrations of diazepam and its metabolites after peroral, intramuscular, and rectal administration. Int. J. Clin. Pharmacol. Biopharm. 12, 427—432. Kapur, J., 2002. Prehospital treatment of status epilepticus with benzodiazepines is effective and safe. Epilepsy Curr. 2, 121—124. Lombroso, C.T., 1989. Intermittent home treatment of status and clusters of seizures. Epilepsia 30 (Suppl. 2), S11—S14. Magnussen, I., Oxlund, H.R.W., Albirk, K.E., Arnold, E., 1979. Absorption of diazepam in man following rectal and parenteral administration. Acta Pharmacol. Toxicol. 45, 87—90. Milligan, N.M., Dhillon, S., Griffiths, A., Oxley, J., Richens, A., 1983. A clinical trial of single dose rectal and oral administration of diazepam for the prevention of serial seizures in adult epileptic patients. J. Neurol. Neurosurg. Psychiatry 47, 235—240. Mitchell, W.G., 1996. Status epilepticus and acute repetitive seizures in children, adolescents, and young adults: etiology, outcome, and treatment. Epilepsia 37 (Suppl. 1), S74—S80. Rey, E., Treluyer, J.M., Pons, G., 1999. Pharmacokinetic optimisation of benzodiazepine therapy for acute seizures. Clin. Pharmacokinet. 36, 409—424. Schroeder, M.C., Wolff, D.L., Maister, B.H., 1996. Lorazepam Intensol® for the management of hospitalized pediatric patients with epilepsy. Epilepsia 37 (Suppl. 5), S154 [abstract]. Tuttle, B.C., 1977. Intramuscular injections and bioavailability. Am. J. Hosp. Pharm. 34, 965—968. Vajda, M., Tegyey, Z., Kozma, M., Vereczkey, L., 1986. Studies on the pharmacokinetics and metabolism of N(4)-carbamoyl-1,3,4,5dihydro-diazepam (Uxepam® ) in rats, dogs and man. J. Pharm. Biomed. Anal. 4, 497—503. July 2005. Vanquix [package insert]. Meridian Medical Technologies, Columbia, MD. Walker, M.C., Tong, X., Brown, S., Shorvon, S.D., Patsalos, P.N., 1998. Comparison of single- and repeated-dose pharmacokinetics of diazepam. Epilepsia 39, 283—289. Yager, J.Y., Seshia, S.S., 1988. Sublingual lorazepam in childhood serial seizures. Am. J. Dis. Child. 142, 931—932.